AssemblyScript implementation of chacha20poly1305, it's 20% - 60% faster compared to stablelib with no memory allocation in the middle.
This was ported from the typescript version https://github.com/StableLib/stablelib/tree/master/packages/chacha20poly1305
yarn add @chainsafe/as-chacha20poly1305
const ctx = newInstance();
const asImpl = new ChaCha20Poly1305(ctx);
const key = new Uint8Array(crypto.randomBytes(KEY_LENGTH));
const nonce = new Uint8Array(crypto.randomBytes(NONCE_LENGTH));
const plainText = new Uint8Array(crypto.randomBytes(512));
const ad = new Uint8Array(crypto.randomBytes(32))
const asSealed = asImpl.seal(key, nonce, plainText, ad);
// overwrite sealed to save memory allocation
const plainText2 = asImpl.open(key, nonce, sealed, true, ad);
expect(plainText2).to.be.deep.equal(plainText);Benchmark results of this implementation (chainSafe), stablelib 1.0.1 and noble 0.2.0 on a Mac M1 as of Aug 2023
chacha20poly1305
✓ chainsafe seal 32 bytes 721249.9 ops/s 1.386482 us/op - 21318 runs 30.1 s
✓ stablelib seal 32 bytes 144494.9 ops/s 6.920660 us/op - 4265 runs 30.0 s
✓ noble seal 32 bytes 150163.8 ops/s 6.659396 us/op - 4434 runs 30.0 s
✓ chainsafe open 32 bytes 671504.2 ops/s 1.489194 us/op - 19851 runs 30.1 s
✓ stablelib open 32 bytes 147999.8 ops/s 6.756768 us/op - 4363 runs 30.1 s
✓ noble open 32 bytes 157093.5 ops/s 6.365635 us/op - 4636 runs 30.0 s
✓ chainsafe seal 64 bytes 494830.5 ops/s 2.020894 us/op - 14610 runs 30.0 s
✓ stablelib seal 64 bytes 136125.4 ops/s 7.346169 us/op - 4025 runs 30.1 s
✓ noble seal 64 bytes 150779.6 ops/s 6.632199 us/op - 4459 runs 30.1 s
✓ chainsafe open 64 bytes 635232.8 ops/s 1.574226 us/op - 18745 runs 30.0 s
✓ stablelib open 64 bytes 145797.0 ops/s 6.858851 us/op - 4308 runs 30.1 s
✓ noble open 64 bytes 148018.9 ops/s 6.755892 us/op - 4371 runs 30.0 s
✓ chainsafe seal 1.00 KB 99285.50 ops/s 10.07196 us/op - 2924 runs 30.0 s
✓ stablelib seal 1.00 KB 38798.05 ops/s 25.77449 us/op - 1149 runs 30.1 s
✓ noble seal 1.00 KB 82057.41 ops/s 12.18659 us/op - 2424 runs 30.0 s
✓ chainsafe open 1.00 KB 100943.3 ops/s 9.906554 us/op - 2982 runs 30.1 s
✓ stablelib open 1.00 KB 36347.17 ops/s 27.51246 us/op - 1073 runs 30.0 s
✓ noble open 1.00 KB 80123.76 ops/s 12.48069 us/op - 2368 runs 30.1 s
✓ chainsafe seal 4.00 KB 28374.74 ops/s 35.24261 us/op - 837 runs 30.1 s
✓ stablelib seal 4.00 KB 11710.30 ops/s 85.39493 us/op - 347 runs 30.2 s
✓ noble seal 4.00 KB 30496.07 ops/s 32.79111 us/op - 900 runs 30.0 s
✓ chainsafe open 4.00 KB 29200.03 ops/s 34.24654 us/op - 861 runs 30.1 s
✓ stablelib open 4.00 KB 10933.00 ops/s 91.46624 us/op - 323 runs 30.1 s
✓ noble open 4.00 KB 30155.13 ops/s 33.16186 us/op - 892 runs 30.1 s
✓ chainsafe seal 8.00 KB 14577.58 ops/s 68.59850 us/op - 431 runs 30.2 s
✓ stablelib seal 8.00 KB 5880.412 ops/s 170.0561 us/op - 174 runs 30.3 s
✓ noble seal 8.00 KB 16014.05 ops/s 62.44518 us/op - 473 runs 30.1 s
✓ chainsafe open 8.00 KB 14990.32 ops/s 66.70972 us/op - 443 runs 30.1 s
✓ stablelib open 8.00 KB 5563.228 ops/s 179.7517 us/op - 165 runs 30.2 s
✓ noble open 8.00 KB 15821.22 ops/s 63.20624 us/op - 467 runs 30.1 s
✓ chainsafe seal 64.00 KB 1784.257 ops/s 560.4575 us/op - 54 runs 30.8 s
✓ stablelib seal 64.00 KB 735.6552 ops/s 1.359332 ms/op - 23 runs 32.6 s
✓ noble seal 64.00 KB 2020.801 ops/s 494.8534 us/op - 60 runs 30.7 s
✓ chainsafe open 64.00 KB 1854.636 ops/s 539.1893 us/op - 56 runs 30.8 s
✓ stablelib open 64.00 KB 691.5201 ops/s 1.446090 ms/op - 21 runs 31.8 s
✓ noble open 64.00 KB 2085.731 ops/s 479.4481 us/op - 62 runs 30.9 s
Apache 2.0